ANPRSmaC: An Automatic Number Plate Recognizer for Intelligent Transportation in Smart City

In recent years, automatic vehicle number plate recognition using stationary cameras are gaining attention due to its various real-life applications such as automatic toll collection, automatic parking system, controlled access in premises, and traffic violations. In past, several approaches for automatic number plate recognition is proposed using conventional methods, machine learning, and more recently using the deep learning. In this paper, we propose an automatic number plate recognizer (ANPRSmaC) to faciliate the intelligent transportation in the smart city. The ANPRSmaC is primarily comprised of four stages such as car detection, number plate localization, character segmentation, and character recognition. In ANPRSmaC, the car detection and number plate localization is accomplished using the YoLo V8 pre-trained model. On the contrary, the we trained a CNN model for the character recognition. The ANPRSmac has used YoLo V5, YoLoV8, and Faster-RCNN to validate the efficiency of the models. The YoLo V5, YoLoV8, and Faster-RCNN have resulted in accuracy of 0.87, 0.994, and 0.992, respectively.